The principal objective of this project is to optimize the use of alternative marrow donors for those patients with severe aplastic anemia or myelodysplasia who lack siblings with an identical HLA genotype. Haploidentical family members incompatible with the patient for one or two HLA antigens of the nonshared haplotype and unrelated volunteers HLA- matched or incompatible with the patient for one HLA antigen will be used as marrow donors. Specific aims are: 1) to test whether cyclophosphamide and ATG will cause sufficient immunosuppression to allow engraftment in severe aplastic patients transplanted from unrelated donors matched for HLA by serology and isolectric focusing of HLA class I molecules and by genotyping of HLA class II; 2) to continue the current phase II study to determine the rates of engraftment, acute graft-versus-host disease (GVHD), and survival in patients with severe aplastic anemia transplanted from HLA-incompatible unrelated or related donors, using conditioning with cyclophosphamide and total body irradiation (TBI) and GVHD prophylaxis with short methotrexate and cyclosporine; 3) to test whether substitution of busulfan for TBI in patients with myelodysplasia without excess blasts can reduce transplant-related mortality without causing an increase in the frequency of graft failure or relapse, and whether a preparative regimen which combines busulfan, cyclophosphamide, and TBI in patients with myelodysplasia with excess blasts can reduce the relapse rate without an increase in toxic deaths; 4) to define efficacy of in vivo treatment with a murine monoclonal antibody specific for the T-cell receptor/CD3 complex to establish and sustain marrow engraftment in patients at high risk for graft rejection because of prior blood transfusions and sensitization to their marrow donors or because of a previous marrow graft rejection; 5) to decrease the risk of bacterial and fungal infection by stimulating earlier recovery of granulocytes with administration of granulocyte, macrophage colony-stimulating factor (GM- CSF) for the first 4 weeks after marrow infusion; 6) to improve treatment of acute and chronic GVHD by intensifying posttransplant immunosuppression. Treatment of acute GVHD with a T-cell-specific immunotoxin (ricin-A chain conjugated to a murine anti-CD5 antibody) and methylprednisolone will be compared to treatment with methylprednisolone alone in a phase III study. 7) Continued treatment with cyclosporine for 720 days will be compared to treatment for 180 days in a phase III study of prevention of chronic GVHD.